多车场异构车型绿色路径规划问题研究  

Multi-Depot Heterogeneous Vehicle Green Routing Problem Research

作　　者：芦立华[1] 姜维 LU Li-hua;JIANG Wei(School of Electronic and Information Engineering,Shanghai Dianji University,Shanghai 201306,China)

机构地区：[1]上海电机学院电子信息学院,上海201306

出　　处：《计算机仿真》2024年第9期281-287,共7页Computer Simulation

基　　金：自然科学基金资助项目(61703267)。

摘　　要：为拓展多车场车辆路径类问题(MDVRP)的求解方式及途径,引入了油耗和碳排放量的近似计算方法,建立了带时间窗的多车场多车型绿色车辆路径问题模型(GMDHF-VRPTW)。将物流配送和碳排放总成本最小化作为优化目标,设计改进的禁忌搜索算法(ITS)求解该模型。为验证提出模型的有效性和可行性,采用Solomon测试数据集,分别从最优目标值、求解时间等方面与经典的禁忌算法进行对比,实验结果表明ITS方法有显著的优势。此外,为降低问题求解的复杂度,结合K-means聚类算法,将多车场转化为单车场,通过判断满足需求的最小车辆数进行建模并提出了基于Gurobi求解器的聚类过滤算法(CF),在求解同规模多车场多车型路径问题进行了探索,结果表明CF颇具优势,能够为企业实际的低碳物流运输提供决策支持和方法指导。In order to solve the variant problems of the Multi-Depot Vehicle Routing Problems(MDVRP),approximate calculation methods for fuel consumption and carbon emissions were introduced,and we established a Green Multi-Depot Heterogeneous Fixed fleet Vehicle Routing Problem model with Time Windows(GMDHF-VRPTW).The optimization objective of our proposed model was to minimize the total cost of freight and carbon emissions.To solve the model,an Improved Tabu Search algorithm(ITS) was designed.To verify the effectiveness and feasibility of the proposed model,the Solomon test dataset was used while the optimal target value and solution time were compared through simulation experiments.The experimental results show that the ITS has significant advantages over the classical tabu search algorithm.Furthermore,considering the problem's complexity,the multi-depots are transformed into a single-depot based on the K-means clustering algorithm.By selecting the minimum number of vehicles to meet the demand,we propose a Clustering Filtering algorithm(CF) based Gurobi solver to simplify the model.CF strategy based on Gurobi has advantages in solving the same scale multi-vehicle and multi-vehicle routing problem,which can provide decision support and method guidance for the actual low carbon logistics transportation of enterprises.

关 键 词：车辆路径问题 绿色物流 多车场异构车 聚类过滤算法 禁忌搜索算法 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]